Investigative Ophthalmology & Visual Science
2000; 41(4). Abstract nr 1185, p. S226
Purpose: To study mechanisms of darkness induction in light of recent models of surface brightness filling-in. Methods: Two subjects judged the brightness of a 1.0 cd/m2, 0.9 deg (radius) disk that was surrounded by two concentric annuli (outer radii = 1.8 and 2.7 deg, respectively) of varying luminance. Disk brightness was measured by adjusting the luminance M of a comparison disk to a achieve a brightness match to the judged disk. The comparison disk was the same size as the judged disk and surrounded by a 5.89 cd/m2, 1.8 deg annulus). Matches were studied as a function of the luminances A1 and A2 of the inner and outer surround annuli, under the constraint that A2 3 A1 3 disk luminance. Results: For each subject, the brightness matches were quantitatively modeled by the following linear regression equation: log M = e1 - e2 log A1 - e3 log2 A1 - e4 log A2 + e5 log A1 log A2. Conclusions: We derive a theoretical brightness matching equation based on a model of darkness induction by a diffusive filling-in process and show that this equation has the same form as the empirical brightness matching equation. The theory is used to extract underlying parameters of the underlying neural filling-in mechanism--such as the percentage of the diffusing surface signal that is stopped by a border as a function of the border contrast--from the coefficients of the brightness matching equation that were estimated from the linear regression analysis of the data.